unity8 is hung in SessionAuthorizer::requestAuthorizationForSession()

Bug #1436860 reported by Albert Astals Cid
6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
Mir
Invalid
Undecided
Unassigned
mir (Ubuntu)
Invalid
Undecided
Unassigned
qtmir (Ubuntu)
Confirmed
Undecided
Unassigned

Bug Description

I was running
    phablet-test-run ubuntuuitoolkit
with the qt dbus patches that fix https://bugs.launchpad.net/ubuntu/+source/unity8/+bug/1421009

at some point the tests failed, i was left on screen with an application that is no longer running, and unity8 seems to be deadlocked

Using mir 0.12.0+15.04.20150228-0ubuntu1

Revision history for this message
Albert Astals Cid (aacid) wrote :
description: updated
Revision history for this message
Albert Astals Cid (aacid) wrote :
Revision history for this message
Albert Astals Cid (aacid) wrote :
Revision history for this message
Gerry Boland (gerboland) wrote :
description: updated
Revision history for this message
Alberto Aguirre (albaguirre) wrote :

Mir 0.12.1 addresses (to be released soon) some race conditions which can prevent a client Mir wait handle from signaling correctly (Thread 9 in the backtrace attached)

https://launchpad.net/mir/+milestone/0.12.1

Changed in mir (Ubuntu):
status: New → Confirmed
Revision history for this message
Daniel van Vugt (vanvugt) wrote :

The server is blocked in QtMir (SessionAuthorizer) trying to authorize a new connection. I've seen this bug reported before. It is a duplicate. I just can't find the original bug report right now...

Thread 22 (Thread 0xae2ff450 (LWP 2300)):

#0 0xffffffff in __libc_do_syscall () at ../sysdeps/unix/sysv/linux/arm/libc-do-syscall.S:46
#1 0xffffffff in __pthread_cond_wait (cond=0xae358b20, mutex=0xae358b08) at pthread_cond_wait.c:186
#2 0xffffffff in QWaitCondition::wait(QMutex*, unsigned long) (time=4294967295, this=0xae358b08) at thread/qwaitcondition_unix.cpp:128
#3 0xffffffff in QWaitCondition::wait(QMutex*, unsigned long) (this=<optimized out>, mutex=mutex@entry=0xae35e160, time=time@entry=4294967295) at thread/qwaitcondition_unix.cpp:200
#4 0xffffffff in QSemaphore::acquire(int) (this=this@entry=0xae2feaf4, n=n@entry=1) at thread/qsemaphore.cpp:137
#5 0xffffffff in QMetaObject::activate(QObject*, int, int, void**) (sender=0xb1a3a064, signalOffset=<optimized out>, local_signal_index=<optimized out>, argv=0xae2feb38) at kernel/qobject.cpp:3685
#6 0xffffffff in SessionAuthorizer::requestAuthorizationForSession(unsigned long long const&, bool&) () at /usr/lib/arm-linux-gnueabihf/qt5/plugins/platforms/libqpa-mirserver.so
#7 0xffffffff in SessionAuthorizer::connection_is_allowed(mir::frontend::SessionCredentials const&) () at /usr/lib/arm-linux-gnueabihf/qt5/plugins/platforms/libqpa-mirserver.so
#8 0xffffffff in mir::frontend::ProtobufConnectionCreator::create_connection_for(std::shared_ptr<boost::asio::basic_stream_socket<boost::asio::local::stream_protocol, boost::asio::stream_socket_service<boost::asio::local::stream_protocol> > > const&, mir::frontend::ConnectionContext const&) (this=0xb1a47f8c, socket=std::shared_ptr (count 2, weak 0) 0xae36d1e4, connection_context=...) at /build/buildd/mir-0.12.0+15.04.20150228/src/server/frontend/protobuf_connection_creator.cpp:65

Changed in mir (Ubuntu):
status: Confirmed → Invalid
Changed in unity8 (Ubuntu):
status: New → Confirmed
affects: unity8 (Ubuntu) → qtmir (Ubuntu)
Changed in qtmir:
status: New → Confirmed
summary: - unity8 is deadlocked
+ unity8 is hung in SessionAuthorizer::requestAuthorizationForSession()
Changed in mir:
status: New → Invalid
Revision history for this message
Daniel van Vugt (vanvugt) wrote :

Found the original report. It was bug 1421308.

Revision history for this message
Albert Astals Cid (aacid) wrote :

Has nothing to do with that one, or either that one was incorrectly duplicated to 1421009

Also it's interesting since updating to Mir 12.1 from silo 13 seems to makes it go away.

Revision history for this message
Daniel van Vugt (vanvugt) wrote :

That one was possibly incorrectly marked as a duplicate. Because this bug shows a hang in an identical location to that of bug 1421308:

Compare comment #6 above with:
https://bugs.launchpad.net/qtmir/+bug/1421308/comments/5

Revision history for this message
Gerry Boland (gerboland) wrote :

This hasn't happened in a while. It is usually due to the Qt GUI loop being blocked.

Michał Sawicz (saviq)
no longer affects: qtmir
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.